Key points are not available for this paper at this time.
As stated in the behavioral model of the environment, the built environment has a clear influence on walking at three different levels: origins and destinations, the areas surrounding them, and the routes connecting them. Nevertheless, the vast majority of research on the relationship between the built environment and travel has focused on origins and the areas surrounding them (the residential neighborhood), so our understanding of the influence of destinations and routes remains limited. In this paper, we develop a structural equations model to explain pedestrian commuting behavior, i.e., walking distance and number of walking trips, controlling for socioeconomic factors, attitudes, and commuting distance, while comparing the built environments of origins with those of destinations. The built environment was described by means of several global information systems (GIS)–based indicators reflecting density, diversity, design, and accessibility, using floating catchment areas for each building identified as an origin or destination of any trip. Our results show that the characteristics of destinations are significant predictors of walking behavior. Moreover, accessibility assumes a mediating role between the built environment and walking behavior, suggesting that it should be explicitly measured to explain that behavior and conceived not as an additional dimension of the built environment, but as a variable determined by it. Therefore, special attention should be paid towards urban planning of major urban destinations, since improving their multimodal accessibility and local walking conditions can contribute to increased walking activity.
Vale et al. (Mon,) studied this question.